Why is it so Hard to Hire Great People

Derek Thompson writes about the difficulty of predicting who will be outstanding at work. He cites Laszlo Bock, senior vice president for Google's people operations, who commented on a study Google had conducted to see whether anyone was good at interviewing, saying,

We looked at tens of thousands of interviews, and everyone who had done the interviews and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship. It's a complete random mess.

Nothing mattered in interviewers' predictions, not even college attended or GPA. 

The only person who was successful at interviewing was someone who interviewed "for a very specialized area, where he happened to be the world's leading expert."

Thompson compares the ability to predict future success at Google to drafting NBA and NFL players. Compared to success in the NBA, a player's success in the NFL depends on a much more complex tangled web of interactions due to the larger number of players on the field. Thus, the draft status of an NBA player more closely corresponds to his success than does the draft status of an NFL player.

The apparent unimportance of university background and GPA in predicting success may also support, as Ulises Mejias speculates, the demise of higher education and the evolution of the job interview into 

a week-long contest where applicants show what they know, what they are capable of doing, and are ranked accordingly. It will include lengthy exams, personality tests, group exercises, creative assignments, detailed simulations, and an army of tests intended to measure everything from know-how to moral character ....

As Mejias admits, such tests may not be accurate. Naturally. Most of the tests are not different from what students do in college. Using simulations is no different than predicting NFL success on the basis of college success. And it's likely that "group exercises" may simulate one level of activity, but not the level of performance in the workplace. 

Still, it will be interesting to see if such week-long "interviews" can be designed and tweaked sufficiently to give substantial predictive results.